OpenAlgo Execution Expert

Knowledge base for building production-grade algorithmic trading strategies on OpenAlgo. Every strategy is a single Python file that toggles between backtest mode (VectorBT) and live execution mode (OpenAlgo SDK + WebSocket) via one CLI flag (--mode backtest|live) or env var (MODE=...).

Strategies are also upload-ready for OpenAlgo's self-hosted /python strategy host.

Core principles

1. One file, two modes. The same signals(df) function feeds both VectorBT (backtest) and the live event loop. Risk thresholds and cost assumptions are honored on both sides.
2. OpenAlgo for everything broker-side. Data via client.history() and WebSocket. Orders via client.placeorder() / placesmartorder() / optionsmultiorder(). Live vs sandbox is decided in OpenAlgo's UI analyzer toggle - the strategy code never knows.
3. Indicator library is user's choice - openalgo.ta (default) or talib. Specialty indicators (Supertrend, Donchian, Ichimoku, HMA, KAMA) always come from openalgo. See rules/indicator-libraries.md.
4. Three execution types - eoc (end-of-candle MARKET), limit (real-time pegged LIMIT), stop (broker-side SL-M trigger). User picks at strategy creation. See rules/execution-types.md.
5. Real-world costs and slippage baked into every backtest (matches vectorbt-backtesting-skills 4-segment Indian model). See rules/transaction-costs.md and rules/slippage-handling.md.
6. Self-hosted /python compatible - every strategy reads env vars in the canonical priority, traps SIGTERM, logs to stdout. See rules/self-hosted-strategies.md.

When to read which rule